INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Nearly two centuries after its introduction, the opiate morphine remains a clinically important drug with major applications for analgesia. A major drawback for the safe clinical use of morphine, however, is that it depresses respiration and can induce hypotension and bradycardia. The effects of opioids on various respiratory parameters depend on the species, route of administration, and the state of the animal, including differences in the anesthetized versus conscious state. Nevertheless, in humans and in rats in the conscious and unconscious state, morphine given at, and in some cases below, analgesic doses is associated with respiratory depression accompanied by decreased sensitivity to CO₂ and hypoxia (1). Most morphine-like compounds at equianalgesic doses have similar respiratory effects (1).

The actions of morphine are primarily mediated through the µ-opioid receptor (MOR), but additional actions can occur at higher doses through interactions with - and -opioid receptors. The receptor interactions involved in the respiratory depressant effects of opiates have long been known to be distinguishable from those involved in analgesia (4). However, the precise role of the various types (µ, , and ) and putative subtypes of opioid receptors in respiratory regulation has remained controversial (1, 2, 5, 6). New compounds that could take advantage of the analgesic effectiveness of activation of the µ-receptor, exhibit high selectivity for that receptor, and show greater separation of analgesic from respiratory effects could be of considerable therapeutic value.

Zadina and colleagues discovered two endogenous µ-opioid receptor agonists: endomorphin 1 (EM1; Tyr-Pro-Trp-Phe-NH₂) and endomorphin 2 (EM2; Tyr-Pro-Phe-Phe-NH₂) (7). The high affinity and selectivity of these peptides for the µ-opioid receptor provide opportunities to explore the effects of specific activation of µ-receptors and the possibility that these ligands may induce a different profile of cardiorespiratory effects from that of morphine or peptide analogs of previously known opioids, such as enkephalin. Relatively dense EMl-like and EM2-like immunoreactivities have been demonstrated in areas associated with high densities of µ-opioid receptors and central cardiorespiratory regulation, including the nucleus tractus solitarius (nTS) and the parabrachial nuclei (8). Systemic injection of these compounds in anesthetized rats elicited dose-dependent hypotensive responses concomitant with decreases in vascular bed resistance (9, 10). However, the cardiovascular and ventilatory effects of EM1 and EM2 in conscious preparations, and their ventilatory effects in unconscious animals, are currently unknown. The present study was undertaken to investigate the cardiorespiratory effects of these newly discovered µ-opioid receptor compounds in relation to their analgesic effects, and to compare them to the enkephalin-derived peptide DAMGO in conscious, freely behaving rats, as well as anesthetized rats.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Adult male Sprague-Dawley rats (280-320 g; Charles River, Wilmington, MA) were used throughout these experiments. The experimental protocols were approved by the Institutional Animal Use and Care Committee. Animals were provided with water and rat chow ad libitum, kept on a light:dark cycle of 12:12 h, and at 22 ± 1° C ambient temperature for at least 1 wk of habituation before surgery and during the postoperative period. For habituation purposes, the animals spent at least 1-2 h of each day in a whole-body plethysmographic chamber.

Chemicals

Endomorphin 1 (EM1; Tyr-Pro-Trp-Phe-NH₂) and endomorphin 2 (EM2; Tyr-Pro-Phe-Phe-NH₂) were custom synthesized by American Peptide Company (Sunnyvale, CA), DAMGO (Tyr-D-Ala-Gly- MePhe-Gly-ol) was obtained from Multiple Peptide Systems (San Diego, CA) through the National Institute on Drug Abuse (NIDA), and morphine sulfate was obtained from Wyeth Pharmaceuticals (Philadelphia, PA). The agonists were dissolved in 0.9% NaCl (pH 7.4) and administered intravenously in 0.5-ml boluses. Naloxone hydrochloride (Nlx, 1 mg/kg; Research Biochemical International, Natick, MA) and naloxone methiodide (MeNlx, 1 mg/kg; Research Biochemical International) were dissolved in 0.9% NaCl and administered intravenously. Control animals received the saline vehicle (0.5 ml of 0.9% NaCl).

Animal Surgery and Preparation

Anesthesia was induced by sodium pentobarbital (Nembutal; 50 mg/ kg, intraperitoneal) and core temperature was measured with a Harvard (South Natick, MA) rectal temperature probe and maintained at 37.5° C via a servo-controlled heating blanket. A 1-cm incision of the left inguinal skin was performed and indwelling polyethylene catheters (PE-50) were inserted into the left femoral artery and vein and advanced to the abdominal aorta and inferior vena cava for drug administration and measurement of cardiovascular parameters. The catheters were secured in the groin and exteriorized in the dorsal aspect of the neck. The vascular lines were flushed with heparinized saline (1,000 U/ml), and the exposed end was heat-sealed and stored in a plastic cap sutured to the skin between the shoulder blades. After surgery, animals were allowed to recover for at least 48 h as demonstrated by the return to normal feeding and sleep-waking patterns.

Ventilatory and Cardiovascular Measurements

Cardiorespiratory responses were continuously measured in the freely behaving, unrestrained animal in a calibrated 3-L open-type barometric chamber (Buxco Electronics, Troy, NY), using methods described by Bartlett and Tenney (11). To minimize the effect of signal drift due to external temperature and pressure changes, a reference chamber of equal size, in which the temperature was measured by a T-type thermocouple, was used. In addition, as previously recommended by Epstein and colleagues, a correction factor was incorporated into the software routine to account for inspiratory and expiratory barometric asymmetries (12). Chamber temperature was maintained slightly below the thermoneutral range (24-28° C). A calibration volume of 0.5 ml of air was repeatedly introduced into the chamber before and on completion of the recordings to monitor chamber calibration. At least 60 min before the start of each protocol, animals were allowed to acclimate to the chamber, in which humidified air (90% relative humidity) was passed through at a rate of 5 L min¹, using a precision flow pump-reservoir system. Pressure changes in the chamber due to the inspiratory and expiratory changes were measured with a high-gain differential pressure transducer (model MP45-1; Validyne, Northridge, CA). Analog signals were continuously digitized and analyzed on-line by a microcomputer software program (Buxco Electronics). A rejection algorithm was included in the breath-by-breath analysis routine and allowed for accurate rejection of motion-induced artifacts. Tidal volume (VT), respiratory frequency (fR), and minute ventilation (E) were computed and stored for subsequent off-line analysis. For clarity, the figures present data as changes in E. As the product of volume and frequency, E provides the most relevant composite measure of respiratory function. In addition, tidal volume was relatively constant under the conditions tested, showing only minor deviations from 1.2 ml. Systemic arterial pressure was measured from the femoral catheter connected to a calibrated pressure transducer via a custom-designed swivel apparatus in the recording chamber (Buxco Electronics). Physiological signals were digitized and a beat-to-beat, peak-trough analysis routine allowed computation of heart rate (HR) and mean systemic arterial blood pressure (mBP, []). On completion of the experimental protocol, animals were euthanized by an intraperitoneal sodium pentobarbital overdose.

Experimental Protocol

The cardioventilatory effects of µ-opioid receptor agonists EM1, EM2, and DAMGO were assessed in the freely behaving rat. After stable cardioventilatory baseline recordings were ascertained in room air, the effects of systemic intravenous injections of EM1, EM2, or DAMGO were assessed and compared with those elicited after administration of 0.5 ml of vehicle. Dose-response curves were generated by intravenous administration of increasing concentrations of EM1 and EM2 (10, 20, 50, 100, 300, 600, 1,200, 2,400, 4,800, and 9,600 nmol/kg), DAMGO (0.5, 1, 5, 10, 20, 50, 100, 300, 600, 1,200, and 2,400 nmol/kg), and morphine sulfate (10, 30, 150, 300, 600, 1,200, 1,500, 1,800, 2,400, 4,800, and 9,600 nmol/kg). Each dosage was administered intravenously with a minimum of 20 min between injections. Dose-response relationships were determined and threshold doses were defined as the lowest doses that elicited changes in cardiorespiratory responses > 1 SD from the mean of the saline baseline for HR, , and E.

In a separate series of experiments, discrimination between central and peripheral sites of opioid action on cardiorespiratory function was assessed by administration of the µ-opioid receptor antagonists naloxone hydrochloride (Nlx), which readily crosses the blood-brain barrier, and naloxone methiodide (MeNlx), which does not cross the blood-brain barrier. The standard effector dose, defined as that eliciting mean E decreases > 2 SD from the mean of saline (control), was determined for EM1 (2,400 nmol/kg), EM2 (2,400 nmol/kg), and DAMGO (600 nmol/kg). These doses were systemically administered after a 30-min baseline period. After complete recovery (60 min), these injections were repeated 10 min after pretreatment with either Nlx or MeNlx (1 mg/kg, intravenous).

Analgesia was tested by the tail-flick assay (13). A 1-in. segment was marked starting 1-in. from the tip of the tail of each rat. This portion of the tail was placed under a focused-light heat source on a standard tail-flick testing apparatus (IITC, Woodland Hills, CA) and the reaction time (withdrawal of the tail) was measured 1 min after opioid administration, the optimal time for measurement in preliminary experiments. The heat intensity was regulated such that the reaction time at baseline was between 3 and 5 s. Analgesia was defined as a latency greater than twice that found before drug administration. A cut off time of 12 s was used to prevent heat-induced blistering. In preliminary experiments, the threshold analgesic dose was determined first with increasing doses of EM1, EM2, and DAMGO (10, 20, 50, 100, 300, 600, 900, 1,200, and 2,400 nmol/kg; n = 8) and morphine sulfate (10, 30, 150, 300, 600, 1,200, 1,500, 1,800, 2,400, 4,800 and 9,600 nmol/kg; n = 8) at 10-min intervals. In a second, separate experiment, the subthreshold (600 nmol/kg) and threshold (900 nmol/kg; n = 8) doses for EM1, EM2, and DAMGO, as well as those for morphine (1,200 and 1,800 nmol/kg; n = 8) were tested. Finally, to assure that the threshold analgesic dose was not the product of drug accumulation, analgesic responses were measured in a third series of experiments after a single threshold dose for each of the compounds. Each of these experiments confirmed the initial determination of the threshold dose. None of the rats were subjected to the heat stimulus more than six times during an experimental period.

Data Analysis

Values are reported as means ± SEM unless indicated otherwise. After a period of adaptation, baseline cardiovascular and ventilatory measurements were determined after saline injection. For each animal, agonist-induced changes from its own baseline were recorded and normalized to percent change. The mean and SD of the baseline values for each group was calculated and normalized to percent change, and the value for significant (threshold) percent change from baseline was defined as ± 1 SD and the standard effector dose as ± 2 SD. Analysis of variance followed by the Newman-Keuls test was used to compare differences among the treatment groups (14). A p value of < 0.05 was set as significant.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Respiratory Responses

Exposure of the rats to the peptide opioid agonists produced a characteristic biphasic ventilatory response with a transient (4- 6 s) E decrease followed by a transient (10-12 min) E increase. In addition, transient decreases in and HR (10-20 s) were observed after opioid agonist treatment (Figure 1). The presence and duration of the effects, however, were dose and drug dependent. Systemic injection of DAMGO elicited the transient E decreases at doses of 10 nmol/kg (n = 8). At higher doses, starting at 100 nmol/kg, an additional excitatory ventilatory component emerged, which characteristically followed the short-lasting ventilatory depression (Figure 1, Table 1A). The duration of this E enhancement was 10-12 min and exhibited dose dependency up to 300 nmol/kg. At this dose, both the early ventilatory reduction and the subsequent ventilatory increase reached a plateau, such that at higher doses no further E changes occurred. Similarly, EM1 and EM2 induced transient E decreases, but the dose of EM1 required for this depression was significantly greater than that of EM2 or DAMGO (Table 1A). The threshold dose of EM1 associated with the transient E decrease was 1,200 versus 100, 10, and 232 nmol/kg for EM2, DAMGO, and morphine, respectively (n = 8, p < 0.01). For DAMGO, EM1, and EM2, but not morphine, E increases developed immediately after the short ventilatory depression. The dose of EM1, and EM2 at which such E enhancements (> 1 SD) occurred was similar to that of DAMGO (100 nmol/kg). Unlike the peptide agonists, morphine failed to elicit an increase in E at any point after administration. E decreases and increases induced by the MOR agonists were primarily mediated by frequency changes while tidal volume was relatively constant under the conditions tested, showing only minor deviations from 1.2 ml.

View larger version (24K): [in this window] [in a new window]   Figure 1.   Representative cardiorespiratory responses showing changes in minute ventilation ( E), mean arterial blood pressure (mBP, [Psa]), and heart rate (HR) over time in the conscious rat after systemic administration of a typical MOR agonist. Intravenous administration of DAMGO (100 nmol/kg) (arrows) induced a biphasic response in E and brief, transient periods of hypotension and bradycardia.

Cardiovascular Responses

As shown in Table 1A, injection of DAMGO at 20 nmol/kg or more was associated with decreases that were extremely transient (1-3 heart beats). Similarly, transient decreases occurred with EM2 starting at 300 nmol/kg, but not with EM1 at any given dose up to 9,600 nmol/kg. In addition, HR decreases usually lasting 10-20 s became evident with DAMGO at 20 nmol/kg (n = 8). This negative chronotropic response also occurred after much higher doses of both EM1 (1,000 nmol/kg, n = 8) and EM2 (1,000 nmol/kg, n = 8) (Table 1A). In addition, morphine induced both transient bradycardic and hypotensive responses to systemic administration. The threshold morphine dose associated with bradycardia was 248 nmol/kg, intravenous (n = 8, p < 0.01), and the dose associated with hypotension was 278 nmol/kg, intravenous (n = 8, p < 0.01) (Table 1A).

Analgesia Responses

In studies examining the threshold dose for analgesia (range, 10-2,400 nmol/kg), the mean doses of DAMGO (n = 12), EM1 (n = 12), EM2 (n = 12), and morphine sulfate (n = 12) needed to double the tail-flick latency response time to a heat stimulus were 900 nmol/kg for all three peptide opioid compounds and 1,725 nmol/kg intravenous for morphine sulfate (Table 1B). However, the ratio of the threshold dose for inducing the E decrease, relative to that for inducing analgesia, differed among all the compounds (Table 1B). Unlike DAMGO, morphine, or EM2, EM1 showed a ratio greater than 1, reflecting a more potent effect on analgesia than on the E decrease.

Opioid Receptor Antagonist Experiments

As shown in Figure 2, intravenous injection of DAMGO (600 nmol/kg, n = 8) induced a significant decrease in E (41.5 ± 7.3%, p < 0.05 versus baseline), followed by a ventilatory increase (39.0 ± 2.9%, p < 0.05 versus baseline). Pretreatment with Nlx (1 mg/kg) abolished the ventilatory reduction induced by DAMGO (11.5 ± 6.2%, p = NS), but MeNlx did not modify this response (41.4 ± 6.4%, p = NS [pre-MeNlx versus post-MeNlx]). The ventilatory increase induced by DAMGO was enhanced after Nlx administration (70.3 ± 21.7%, p < 0.05 versus pre-Nlx), but remained unchanged after MeNlx (39.6 ± 17.5%, p = NS [pre-MeNlx versus post-MeNlx]).

View larger version (9K): [in this window] [in a new window]   Figure 2.   Effects of systemic administration of EM1 (2,400 nmol/kg, n = 8), EM2 (2,400 nmol/kg, n = 8), and DAMGO (600 nmol/kg, n = 8) before (open columns) or after treatment with naloxone (solid columns; 1 mg/kg, intravenous) or methyl-naloxone (hatched columns; 1 mg/kg, intravenous) on percent changes in ventilation ( E) in the conscious rat. The respiratory response to the agonists was biphasic, with an initial decrease followed by an increased E. For each panel, columns depict the maximal decrease (left four columns) and maximal increase (right four columns) in E during the 20-min test period. *p < 0.05, relative to peptide alone.

As with DAMGO, EM2 (2,400 nmol/kg) induced E decreases, which were markedly inhibited by Nlx (32.8 ± 5.6% pre-Nlx versus 3.3 ± 4.0% post-Nlx treatment, p < 0.02). Similarly, Nlx enhanced the E increases elicited by EM2 (35.9 ± 7.4 versus 73.7 ± 22.7%, p < 0.05). Pretreatment with MeNlx failed to modify either E decreases or E increases associated with EM2 injection. When EM1 was given (2,400 nmol/ kg, intravenous), however Nlx reversed the E depression, no effects on E enhancements occurred (Figure 2). Neither component of the E response to EM1 was modified by pretreatment with MeNlx (Figure 2).

DAMGO (600 nmol/kg) induced HR decreases (13.7 ± 2.8%), which were abolished by Nlx (6.8 ± 1.5%, p < 0.05, n = 8; Figure 3). Similarly, in a separate experiment, the bradycardic responses elicited by DAMGO injection (11.5 ± 0.8%) were also reversed by pretreatment with MeNlx (8.8 ± 1.8%, p < 0.05; Figure 3). Similar attenuations of the negative chronotropic responses to EM1 (2,400 nmol/kg, n = 8) and EM2 (2,400 nmol/kg, n = 8) occurred after pretreatment with Nlx and MeNlx (Figure 3). The transient blood pressure decreases associated with injection of the µ-opioid receptor agonists were completely abolished by both Nlx and MeNlx.

View larger version (10K): [in this window] [in a new window]   Figure 3.   Effects of systemic administration of EM1 (2,400 nmol/kg, n = 8), EM2 (2,400 nmol/kg, n = 8), and DAMGO (600 nmol/kg, n = 8) before (open columns) or after treatment with naloxone (solid columns; 1 mg/kg, intravenous) or methyl-naloxone (hatched columns; 1 mg/kg, intravenous) on percent changes in heart rate (HR) in the conscious rat. The decrease in heart rate induced by each of the agonists (open columns) was reversed by both of the antagonists (solid and hatched columns). *p < 0.05, relative to peptide alone.

Anesthesia

The effects of EM1, EM2, and DAMGO obtained in the conscious, freely behaving rats are significantly different from those observed earlier in anesthetized rat preparations (9, 10). We therefore examined the cardiorespiratory responses to intravenous administration of EM1, EM2, and DAMGO in rats anesthetized with sodium pentobarbital (50 mg/kg, intraperitoneal). An adequate plane of anesthesia, as measured by the absence of corneal reflexes, was maintained with additional doses of sodium pentobarbital as previously described (9, 10). Intravenous administration of EM1 (50 nmol/kg, n = 4), EM2 (50 nmol/kg, n = 4), and DAMGO (5 nmol/kg, n = 4) elicited marked hypotension and bradycardia. Both cardiovascular responses were rapid in onset and exhibited prolonged recovery times (Figure 4A). In addition, immediately after intravenous administration of each MOR agonist, an initial short-lasting apnea followed by a resumption of baseline E was observed. To confirm our results in the unanesthetized model, each rat was allowed 24 h of recovery and then the experimental protocol was repeated in the now conscious and free behaving rat. Intravenous administration of EM1, EM2, and DAMGO, at doses eliciting changes in cardioventilatory function in the anesthetized rat, failed to induce significant changes in the conscious, freely behaving rat (Figure 4B). At higher doses in the conscious rat, however, changes in cardioventilatory functioning were observed as described above.

View larger version (39K): [in this window] [in a new window]   Figure 4.   EM1-induced changes in minute ventilation ( E), mean arterial blood pressure (mBP, [Psa]), and heart rate (HR) in (A) a rat anesthetized with sodium pentobarbital (50 mg/kg, intraperitoneal) compared with (B) an absence of cardioventilatory changes after the same dose of EM1 in an unanesthetized, conscious, and freely behaving rat. Arrows indicate intravenous administration of EM1 (50 nmol/kg). Note that the time scale for E in (A) has been expanded to illustrate the brief period of apnea (dotted line) followed by a resumption of baseline E after administration of EM1.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

At analgesic doses in humans, systemic morphine administration depresses respiration, primarily because of a decrease in respiratory rate (1, 2). Previous research supports a role for µ-opiod receptors in opioid-induced respiratory depression (1, 2, 5, 6, 15, 16). In conscious rats tested in the present study, morphine at low doses induced a transient E decrease, but unlike the peptide MOR agonists tested, morphine showed no subsequent E increase. EM1, EM2, and DAMGO, in contrast, exerted a dual effect: a rapid ventilatory depression followed by increased ventilation. These ventilatory responses were accompanied by a transient bradycardia and short-lasting hypotension. The cardiovascular effects were apparently mediated, at least in part, by peripheral mechanisms because they were blocked by an opiate antagonist (MeNlx) restricted to the periphery. In contrast, the ventilatory reduction was abolished only by an antagonist capable of acting both centrally and peripherally (Nlx), indicating a central action of the peptide on the respiratory function.

Widespread distribution of µ-opioid receptor mRNA expression occurs in the brainstem, including in areas associated with respiratory function such as the nucleus tractus solitarius (nTS), the parabrachial (PB) nucleus, the nucleus ambiguus, and the rostrocaudal ventral medullary surface (17). In addition, these areas are among the structures containing the highest density of EM1-like and EM2-like immunoreactivities (EM1-LI and EM2-LI, respectively) in the CNS, indicating a natural role for these peptides in regulating respiratory and cardiovascular function. The expression pattern of EM1-LI and EM2-LI, however, differs among these brainstem nuclei (8), a difference that may be related to the different effects of the peptides at submaximal doses on cardiorespiratory modulation. EM1-LI is prominent, for example, in the dorsomedial nTS and the external lateral PB nuclei, which are particularly important for cardiovascular regulation. In contrast, EM2-LI is more prominent in the ventrolateral nTS and external medial PB and Kölliker Fuse nuclei, which are crucial regions for respiratory regulation (18). EM2 was more likely to induce respiratory depression at lower doses compared with EM1. Evidence is accumulating that EM1 and EM2 have subtle differences in their cellular actions. Like the first report on endomorphins (7), many studies have reported greater potency of EM1 than EM2. These include the demonstration of different effects of the two peptides on the firing frequency of medullary neurons associated with cardiovascular function (19). Apparent qualitative differences in the cellular actions of EM1 and EM2 include differential inhibition of dorsal horn neuron activation by primary afferents (20), and differences in sensitivity to naloxonazine (21). Sanchez-Blazquez and colleagues, using antisense oligodeoxynucleotide probes targeting different G protein subunits (22) or different exons of the µ-receptor (23), differentially blocked the antinociceptive effects of EM1, EM2, DAMGO, and morphine. Therefore, the differences among these compounds in their potency and physiological effects may result from differences in their respective structures, binding, or signaling characteristics, and, for EM1 and EM2, their distribution.

The mechanisms underlying the excitatory ventilatory responses to MOR agonists are currently unclear. Because the response was not attenuated by either of the antagonists (Nlx and MeNlx), it would appear that this component is not opioid mediated. Indeed, after injection of DAMGO and EM2, naloxone enhanced the excitatory ventilatory response, indicating an opioid-induced inhibition. The excitatory component therefore appears to be a compensatory reaction to the effects of the opioids, and the opioids may suppress this component either directly or by release of endogenous opioids. Treatment with Nlx, but not MeNlx, "unmasks" this inhibition, indicating that central actions of opioids can inhibit the ventilatory excitation.

Of particular interest was the finding that EM1 showed the most robust excitatory E response, which was not enhanced by Nlx or MeNlx. The E increase after DAMGO and EM2 administration reached the level of the EM1-induced increase only after enhancement by naloxone. Thus, although equivalent respiratory depression was achieved with the doses used, the apparently compensatory E excitatory component was inhibited by DAMGO and EM2, but not by EM1. Two measures therefore indicate that EM1 is less prone to induce respiratory depression than DAMGO or EM2. A higher dose is required to produce the initial E decrease, and even at doses that produce equivalent E decreases, EM1 does not produce a naloxone-reversible inhibition of the subsequent excitatory E component while EM2 and DAMGO do. Because there is evidence of MOR expression within the carotid bodies (24), the possibility exists that the excitatory response to MOR agonists could be mediated by peripheral mechanisms exerting a tonic effect on ventilatory output. However, the experiments using MeNlx indicate that this is not the case. Thus, we postulate that the respiratory increases elicited by EM1, EM2, and DAMGO originated centrally rather than peripherally.

In the present study, systemic administration of EM1 and EM2 induced transient bradycardia and hypotension in the conscious rat. Each of these responses was blocked by naloxone (Nlx) and its quaternary analog MeNlx, suggesting that the endomorphin effect on such cardiovascular measures is primarily mediated via peripheral mechanisms. In agreement with our findings, Kwok and Dun reported that in anesthetized rats, EM1-induced changes in HR and blood pressure were blocked by pretreatment with naloxone, atropine methylnitrate, atropine sulfate, and bilateral vagotomy (25). In addition, systemic administration of EM1 and EM2 decreased cardiac output and total peripheral resistance, as well as hindlimb vascular resistance in the anesthetized rat preparation (9, 10). Collectively, these studies would suggest that the major effects on cardiovascular function by EM1 and EM2 were primarily mediated by interactions with peripherally located µ-opioid receptors. However, we cannot exclude the possibility that a central mechanism is involved, as µ-opioid receptors are highly expressed within regions traditionally considered to be associated with cardiovascular control (26). In addition, although MeNlx is considered to be a compound that does not gain entrance to the brain across the blood-brain barrier, this assumption has been occasionally disputed. For example, Brown and Goldberg (27) have suggested that MeNlx could partially cross the blood-brain barrier in regions such as in the area postrema, to block MORs. The seemingly absent pharmacological effect could also be due to a relatively low affinity of MeNlx for opiate receptors. Notwithstanding such considerations, the changes in HR and elicited by systemic endomorphin administration appear to be primarily mediated through a peripheral mechanism while the effects on respiration are centrally mediated.

Intravenous administration of EM1, EM2, and DAMGO induced a biphasic ventilatory change in the conscious rat. In contrast, however, in the anesthetized rat, intravenous administration of the endomorphins and DAMGO induced a brief apnea and prolonged periods of marked hypotension and bradycardia. These disparate results suggest that the cardioventilatory responses elicited by the endomorphins are dependent on concurrent anesthetic treatment, such that the anesthesia appears to mask an excitatory respiratory component present during consciousness. Our results are consistent with reports suggesting that cardiorespiratory responses induced by MOR agonists are highly dependent on the presence or absence of anesthesia and the route of administration.

Development of analgesic drugs that would be either equipotent or even more potent than morphine, but void of the untoward cardiorespiratory side effects, is of obvious importance. In the original description of EM, intracerebroventricular injection of EM1 in mice induced a potent and prolonged analgesia comparable to that of morphine. In the present study, despite the fact that DAMGO, EM1, and EM2 have similar affinities for the µ-opioid receptor (7) and induced analgesia at similar doses, the doses at which the cardiorespiratory side effects occurred differed greatly. Indeed, the ventilatory depressant effects required significantly higher doses of EM1 than of EM2 or DAMGO. For EM1, the dose inducing ventilatory depression was above that inducing analgesia, while for DAMGO and EM2, ventilatory depression was seen at doses below the analgesic dose. In addition, the transient decrease in blood pressure seen after administration of doses as low as 20 nmol/kg (DAMGO) or 300 nmol/kg (EM2) was not observed with EM1 at doses up to 9,600 nmol/kg. Thus, the results support the concept that the analgesic, respiratory, and cardiovascular effects of µ-agonists can be dissociated and that EM1 shows several characteristics that could provide the basis for novel, safer analgesics.